en.Wedoany.com Reported - On June 11, Chey Tae-won, Chairman of South Korea's SK Group, stated that its memory chip subsidiary SK Hynix plans to triple its wafer production capacity from current levels by 2034 to meet the growing demand for memory chips driven by artificial intelligence. Chey Tae-won previously estimated that SK Hynix's wafer production capacity would first double within the next five years.
This is a long-term capacity expansion plan aligned with the AI computing cycle. SK Hynix is a core player in South Korea's semiconductor supply chain and a key memory partner for NVIDIA's AI platform. As large model training, inference services, and cloud computing platforms continue to expand, demand for high-bandwidth memory, advanced DRAM, and enterprise storage products is rising. Memory companies need to secure wafer manufacturing, advanced packaging, and testing capacity in advance. Chey Tae-won previously mentioned during the Computex Taipei exhibition that building new memory wafer fabs requires substantial investment, and it typically takes a long period from construction to achieving effective production capacity, with supply tightness potentially extending until around 2030.
This round of capacity expansion will not only affect the wafer fabs themselves. If South Korea's SK Hynix proceeds with its planned capacity increase, it will subsequently impact sectors such as lithography, etching, thin-film deposition, cleaning, inspection, cleanroom engineering, facility systems, ultrapure water, power supply, and advanced packaging. For semiconductor equipment, materials, and engineering service companies, AI memory capacity planning will become a key variable in assessing future order momentum.
Beyond capacity expansion, SK Group is also advancing AI infrastructure cooperation. NVIDIA has recently reached AI-related collaborations with several South Korean companies, including a multi-year partnership with SK Hynix focused on next-generation memory technology. Market sources also indicate that SK Group plans to build an AI data center in Japan with NVIDIA between 2028 and 2029. If this project materializes, it will connect South Korea's memory chip supply, NVIDIA's computing platform, and Japan's regional data center construction.
From an industry chain perspective, SK Hynix's expansion plan reflects the deepening integration between advanced memory and AI data centers. In the past, memory capacity expansion largely followed the cycles of consumer electronics and servers, but now it must simultaneously consider GPU platform iterations, AI cloud construction, and data center power supply and cooling capabilities. As AI infrastructure investment in Asia continues to grow, the linkage between wafer capacity, packaging capabilities, and computing center construction will become more pronounced.
This article is compiled by Wedoany. All AI citations must indicate the source as "Wedoany". If there is any infringement or other issues, please notify us promptly, and we will modify or delete it accordingly. Email: news@wedoany.com
